Automated fMRI Feature Abstraction using Neural Network Clustering Techniques

In this paper we propose a method to automatically find useful abstractions of the fMRI data using a new neural network clustering technique. The purpose of these data abstractions is to alleviate the computational burden by reducing dimensionality, to minimize the risk of overfitting by reducing the number of free model parameters, and to uncover what relationships among voxels can help explain in what cognitive state a subject is at a given point in time. We show that our method outperforms classical machine learning methods like SVM, GNB and kNN in terms of accuracy.